How to extract a specific portion of a string in bigquery

Question

I have a string that looks something similar to this:

28f1e5f7-47c6-4d67-bcbf-9e807c379076-9780-gThGFkHY0CeFCPwA6Efys7

I would like to split it based on '-' but also tell it at what position or point to do the split. My goal is to drop the last two strings

The end result of what I want should look something like this:

28f1e5f7-47c6-4d67-bcbf-9e807c379076

I know i can split offset and then concat. I'm wondering if there is a more straight forward way around as it will require me to split 5 times and then concatenate 5 times?

Answer 1

Can you try this Query?

with myTable as (
  select 1 as id, "28f1e5f7-47c6-4d67-bcbf-9e807c379076-9780-gThGFkHY0CeFCPwA6Efys7" as sampleString
  union all select 2 as id, "28f1e5f7-47c6-4d67-bcbf-9e807c379076-9780-gThGFkHY0CeFCPwA6Efys7-dhushkdha" as sampleString
)


SELECT id, sampleString,
  (
    SELECT STRING_AGG(samp, '-' ORDER BY index) 
    FROM UNNEST(SPLIT(sampleString, '-')) samp WITH OFFSET index 
    WHERE index BETWEEN 0 AND (select array_length(split(sampleString,'-')) - 3)
  ) 
FROM myTable

This query will omit the last two values given that it is delimited by "-"

Output:

I added anonther value to the 2nd column just to test if will delete the last two values.

Answer 2

You might consider using INSTR() as well

INSTR(source_value, search_value[, position[, occurrence]])

WITH sample_data AS (
  SELECT '28f1e5f7-47c6-4d67-bcbf-9e807c379076-9780-gThGFkHY0CeFCPwA6Efys7' str
)
SELECT SUBSTR(str, 1, INSTR(str, '-', -1, 2) - 1)  -- drop the last two strings
  FROM sample_data;

Answer 3

yet another few options

select str, 
  regexp_extract(str, r'^(?:[^-]+-){4}[^-]+') first_five_v1,
  substring(str, 1, regexp_instr(str, r'-', 1, 5) - 1) first_five_v2
from your_table           

       

with output